Skillful forecasting of offshore winds from satellite scatterometer constellations

📄 arXiv: 2607.27152v1 📥 PDF

作者: Francesco Pinto, Luca Lanzilao, Paco Lopez Dekker, Angela Meyer

分类: cs.LG

发布日期: 2026-07-29

💡 一句话要点

提出WindCastNet以解决海上风速短期预测问题

🎯 匹配领域: 支柱八：物理动画 (Physics-based Animation)

关键词: 卫星散射计 海上风速预测 短期天气预报 机器学习 长短期记忆网络 可再生能源 气象数据

📋 核心要点

  1. 现有的数值天气预报方法在分钟到小时的短期预测中，初始条件的准确性限制了预测能力。
  2. WindCastNet通过学习不规则的卫星观测数据，提出了一种新的海上风速和风向即时预测框架。
  3. 在北海的实验中，WindCastNet显著降低了预测误差，展示了其在短期海上风预测中的优势。

📝 摘要（中文）

准确的海上风速短期预测对于电力系统运行和日益增长的海上风能整合至关重要。现有的数值天气预报方法在分钟到小时的预测时效上存在局限。本文提出WindCastNet，这是首个基于卫星散射计的海上风速和风向即时预测框架，能够从不规则的卫星观测数据中学习。WindCastNet利用部分卷积长短期记忆网络，处理来自欧洲、中国和印度散射计的微波雷达观测，尽管其空间覆盖不均、采样不一致且重访时间可变。通过在北海的评估，WindCastNet在1小时和2小时的预测中分别减少了23%和7%的均方根误差，且在前3小时的预测中优于持久性模型9-15%。

🔬 方法详解

问题定义：本文旨在解决现有数值天气预报在短期海上风速预测中的不足，尤其是在分钟到小时的预测时效上，初始条件的准确性对预测能力的影响显著。

核心思路：WindCastNet通过直接利用卫星散射计的观测数据，构建了一个新的即时预测框架，能够处理不规则的时空数据，从而提升短期预测的准确性。

技术框架：WindCastNet采用部分卷积长短期记忆网络（PCLSTM），通过编码空间观测掩码和观测间隔，利用连续的时间表示实现任意预测时效的能力。

关键创新：该研究的核心创新在于首次将卫星散射计观测直接应用于短期风速预测，突破了传统数值天气预报的局限，提供了独立且具有竞争力的预测来源。

关键设计：WindCastNet的设计包括处理不规则空间覆盖的能力、异步采样的适应性以及可变重访时间的处理，确保了模型在不同条件下的稳定性和准确性。具体的损失函数和网络结构细节在论文中进行了详细描述。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

WindCastNet在北海的实验结果显示，其在1小时和2小时的预测中分别减少了23%和7%的均方根误差，相较于HARMONIE MEPS模型，且在前3小时的预测中优于持久性模型9-15%，展现了显著的性能提升。

🎯 应用场景

该研究的潜在应用领域包括海上风能的短期预测，为电力系统的优化调度提供支持。此外，WindCastNet还可应用于更广泛的海洋天气预报，如热带气旋的即时预测，具有重要的实际价值和未来影响。

📄 摘要（原文）

Accurate intraday forecasts of offshore wind are becoming increasingly important for power system operation and the integration of growing shares of offshore wind energy. Operational forecasts rely predominantly on numerical weather prediction (NWP), which is not optimized for lead times of minutes to hours, where initial-condition accuracy dominates forecast skill. Although satellite scatterometer observations are routinely assimilated into NWP, they have not previously been used directly for forecasting. Here we present WindCastNet, the first satellite-based nowcasting framework for offshore wind speed and direction, introducing a new paradigm for intraday forecasting that learns from spatiotemporally irregular satellite observations. WindCastNet predicts offshore wind fields from observations acquired by satellite scatterometer constellations. WindCastNet employs a partial convolutional long short-term memory network that exploits microwave radar observations from the European, Chinese, and Indian scatterometers despite their irregular spatial coverage, asynchronous sampling, and variable revisit times. Spatial observation masks and inter-observation intervals are encoded, while a continuous temporal representation enables forecasts at arbitrary lead times. Evaluated over the North Sea, WindCastNet reduces the root-mean-square error by 23% and 7% relative to the HARMONIE MEPS model at lead times of 1 and 2 h, respectively, and outperforms persistence by 9-15% during the first three forecast hours. Forecast skill decreases under strong-wind conditions and spatially non-uniform flow. These results demonstrate that satellite scatterometer constellations can provide an independent and competitive source of short-term offshore wind forecasts, opening new opportunities for renewable energy forecasting but also broader marine weather applications, including tropical cyclone nowcasting.